METHODOLOGICAL STUDIES OF TRACKING BEHAVIOR: THE EFFECT OF VARIOUS SUPPLEMENTAL INFORMATION FEEDBACKS.
Abstract
An experiment was run to explore the use of a direct way of shaping the behavior of human trackers through the use of a visually displayed signal which was a function of the scoring criterion. It had been expected that with a complex criterion, a time-weighted one for example, it would be difficult for a tracker to remember exactly what was desired of him at various times, and that a supplemental way of presenting the information would facilitate his task. The supplemental information was provided by a computed 'criterion circle' which surrounded the error display and whose diameter varied as a function of the computed criterion. Two criteria were used: a simple 'average error' and a time-weighted error. Both performance scores, in terms of the two criteria, and Bode plots were computed after initial training was completed. The on-line information improved the performance on the simple criterion but, contrary to expectation, reduced performance on the complex one. On the basis of the Bode plots, it appears that the supplemental information required attention of the tracker and interfered with his primary task performance. Apparently any added non-redundant visual signal interferes with the performance on the main task. On the other hand, a properly stated set of instructions can be applied by the subject without cost. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 13, 1966
- Accession Number
- AD0649574
Entities
People
- Jane L. Ward
- John W. Senders
Organizations
- BBN Technologies